1 files changed, 28 insertions, 84 deletions
diff --git a/scalding/src/test/scala/sandcrawler/ScorableTest.scala b/scalding/src/test/scala/sandcrawler/ScorableTest.scala
index 40801a0..2f80492 100644
--- a/scalding/src/test/scala/sandcrawler/ScorableTest.scala
+++ b/scalding/src/test/scala/sandcrawler/ScorableTest.scala
@@ -9,7 +9,7 @@ import org.scalatest._
 import parallelai.spyglass.hbase.HBaseConstants.SourceMode
 
 class ScorableTest extends FlatSpec with Matchers {
-      val JsonString = """
+  val JsonString = """
 {
   "title": "<<TITLE>>",
   "authors": [
@@ -55,96 +55,40 @@ class ScorableTest extends FlatSpec with Matchers {
 }
 """
 
-  performUnitTests()
-  performPipelineTests()
-
-  def performUnitTests() {
-    "titleToSlug()" should "extract the parts of titles before a colon" in {
-      Scorable.titleToSlug("HELLO:there") shouldBe "hello"
-    }
-
-    it should "extract an entire colon-less string" in {
-      Scorable.titleToSlug("hello THERE") shouldBe "hello there"
-    }
-
-    it should "return Scorable.NoSlug if given empty string" in {
-      Scorable.titleToSlug("") shouldBe Scorable.NoSlug
-    }
-
-    it should "return Scorable.NoSlug if given null" in {
-      Scorable.titleToSlug(null) shouldBe Scorable.NoSlug
-    }
-
-    "titleToSlug()" should "strip punctuation" in {
-      Scorable.titleToSlug("HELLO!:the:re") shouldBe "hello"
-      Scorable.titleToSlug("a:b:c") shouldBe "a"
-      Scorable.titleToSlug(
-        "If you're happy and you know it, clap your hands!") shouldBe "if youre happy and you know it clap your hands"
-    }
+  "titleToSlug()" should "extract the parts of titles before a colon" in {
+    Scorable.titleToSlug("HELLO:there") shouldBe "hello"
+  }
 
-    "jsonToMap()" should "return a map, given a legal JSON string" in {
-      Scorable.jsonToMap(JsonString) should not be (None)
-    }
+  it should "extract an entire colon-less string" in {
+    Scorable.titleToSlug("hello THERE") shouldBe "hello there"
+  }
 
-    it should "return None, given illegal JSON" in {
-      Scorable.jsonToMap("illegal{,json{{") should be (None)
-    }
+  it should "return Scorable.NoSlug if given empty string" in {
+    Scorable.titleToSlug("") shouldBe Scorable.NoSlug
+  }
 
-    "computeOutput()" should "return Scorable.MaxScore if given identical ReduceFeatures" in {
-      val score = Scorable.computeSimilarity(
-        new ReduceFeatures(JsonString), new ReduceFeatures(JsonString))
-      score shouldBe Scorable.MaxScore
-    }
+  it should "return Scorable.NoSlug if given null" in {
+    Scorable.titleToSlug(null) shouldBe Scorable.NoSlug
   }
 
-  def performPipelineTests() {
-      /*
+  "titleToSlug()" should "strip punctuation" in {
+    Scorable.titleToSlug("HELLO!:the:re") shouldBe "hello"
+    Scorable.titleToSlug("a:b:c") shouldBe "a"
+    Scorable.titleToSlug(
+      "If you're happy and you know it, clap your hands!") shouldBe "if youre happy and you know it clap your hands"
+  }
 
-    val output = "/tmp/testOutput"
-    val input = "/tmp/testInput"
-    val (testTable, testHost) = ("test-table", "dummy-host:2181")
+  "jsonToMap()" should "return a map, given a legal JSON string" in {
+    Scorable.jsonToMap(JsonString) should not be (None)
+  }
 
-  val grobidSampleData = List(
-    List(Bytes.toBytes("sha1:K2DKSSVTXWPRMFDTWSTCQW3RVWRIOV3Q"),
-      Bytes.toBytes(GrobidString.replace("<<TITLE>>", "Title 1"))),
-    List(Bytes.toBytes("sha1:C3YNNEGH5WAG5ZAAXWAEBNXJWT6CZ3WU"),
-      Bytes.toBytes(GrobidString.replace("<<TITLE>>", "Title 2: TNG"))),
-    List(Bytes.toBytes("sha1:SDKUVHC3YNNEGH5WAG5ZAAXWAEBNX4WT"),
-      Bytes.toBytes(GrobidString.replace("<<TITLE>>", "Title 3: The Sequel"))),
-    List(Bytes.toBytes("sha1:35985C3YNNEGH5WAG5ZAAXWAEBNXJW56"), 
-      Bytes.toBytes(MalformedGrobidString)))
+  it should "return None, given illegal JSON" in {
+    Scorable.jsonToMap("illegal{,json{{") should be (None)
+  }
 
-  JobTest("sandcrawler.HBaseCrossrefScoreJob")
-    .arg("test", "")
-    .arg("app.conf.path", "app.conf")
-    .arg("output", output)
-    .arg("hbase-table", testTable)
-    .arg("zookeeper-hosts", testHost)
-    .arg("crossref-input", input)
-    .arg("debug", "true")
-    .source[Tuple](HBaseCrossrefScore.getHBaseSource(testTable, testHost),
-      grobidSampleData.map(l => new Tuple(l.map(s => {new ImmutableBytesWritable(s)}):_*)))
-    .source(TextLine(input), List(
-      0 -> CrossrefString.replace("<<TITLE>>", "Title 1: TNG").replace("<<DOI>>", "DOI-0"),
-      1 -> CrossrefString.replace("<<TITLE>>", "Title 1: TNG 2").replace("<<DOI>>", "DOI-0.5"),
-      2 -> CrossrefString.replace("<<TITLE>>", "Title 1: TNG 3").replace("<<DOI>>", "DOI-0.75"),
-      3 -> CrossrefString.replace("<<TITLE>>", "Title 2: Rebooted").replace("<<DOI>>", "DOI-1")))
-    .sink[(Int, String, String, String, String)](TypedTsv[(Int,
-    String, String, String, String)](output)) {
-      // Grobid titles: 
-      //   "Title 1", "Title 2: TNG", "Title 3: The Sequel"
-      // crossref slugs: 
-      //   "Title 1: TNG", "Title 1: TNG 2", "Title 1: TNG 3", "Title 2 Rebooted"
-      // Join should have 3 "Title  1" slugs and 1 "Title 2" slug
-      outputBuffer =>
-      "The pipeline" should "return a 4-element list" in {
-        outputBuffer should have length 4
-      }
-    }
-    .run
-    .finish
-}
-       */
+  "computeOutput()" should "return Scorable.MaxScore if given identical ReduceFeatures" in {
+    val score = Scorable.computeSimilarity(
+      new ReduceFeatures(JsonString), new ReduceFeatures(JsonString))
+    score shouldBe Scorable.MaxScore
   }
 }
-